CN115086806A

CN115086806A - GPON weak light fault positioning method and system based on classification algorithm

Info

Publication number: CN115086806A
Application number: CN202210226385.8A
Authority: CN
Inventors: 蒋燕; 韩玉琪; 何丹; 张梓豪; 李兵
Original assignee: Guangdong Vocational College Of Post And Telecom
Current assignee: Guangdong Vocational College Of Post And Telecom
Priority date: 2022-03-09
Filing date: 2022-03-09
Publication date: 2022-09-20

Abstract

The invention relates to a GPON weak light fault positioning method and system based on a classification algorithm, and the technical scheme key points are as follows: the method comprises the following steps: collecting and converging ONU optical power data; constructing a training sample by combining optical power data of the ONU, network resource data and historical fault data; constructing a GPON weak light fault positioning model based on a classification algorithm by using the training sample; analyzing the ONU optical power data to obtain fault section positioning data by applying the GPON weak light fault positioning model based on the classification algorithm; the method and the device have the advantages of improving the positioning and scheduling processing efficiency of the fault.

Description

GPON weak light fault positioning method and system based on classification algorithm

Technical Field

The invention relates to the technical field of communication network management, in particular to a GPON weak light fault positioning method and system based on a classification algorithm.

Background

With the rapid development of the home broadband service and the increasing number of users, higher requirements are put forward on the network quality and the maintenance efficiency of the home broadband. The mainstream broadband access technology is the GPON technology, and the weak light of the GPON network is a common problem in maintaining the home wide network. The GPON network mainly comprises an OLT, an ODN and an ONU, and the weak light of the GPON network is mainly reflected in the weak light of the ONU, namely the receiving optical power of the ONU is smaller than the receiving sensitivity of the ONU, so that the problems of instability, frequent disconnection, low network speed and the like of a user network are caused. Therefore, the method has important significance for quickly positioning and processing the GPON weak light fault.

The GPON network generally consists of a large number of optical fibers and passive devices, and network elements such as the ODN often become network monitoring blind spots and are difficult to directly position. At present, fault processing is mainly carried out by means of step-by-step troubleshooting from an OLT to a first-stage optical splitter, a second-stage optical splitter and an ONU by maintenance personnel, and the problems of high cost, low efficiency and the like exist. CN112822128A provides a PON system message mirroring method and a PON system, which can further obtain multiple PON interaction messages in addition to two-layer network messages, and can assist in locating ODN link side faults of a central office and a terminal device of the PON system. CN108810671A provides a method for collecting ONU basic data (physical link state, authentication state, etc.) and reporting the ONU basic data to the OLT based on the original OMCI management channel of the GPON system, which solves the problem that it is difficult to collect relevant information when the ONU cannot communicate with the network management platform, and improves the failure processing efficiency and accuracy of the on-line device. In general, the methods improve the fault location processing efficiency by collecting messages, ONU basic data and other information for analysis and processing.

However, the method collects and analyzes the ONU messages of all types one by one, has huge deployment and analysis costs, and is difficult to be completely spread and applied in a large-scale communication network. In addition, a large amount of labor cost is consumed in the traditional method for checking, analyzing and positioning step by step through maintenance personnel, different fault points in the network are generally processed by different maintenance teams, and fault work orders need to be sent to a plurality of maintenance teams simultaneously when primary positioning does not exist, so that a large amount of maintenance labor resources are wasted, and therefore, a space needs to be improved.

Disclosure of Invention

Aiming at the defects in the prior art, the invention aims to provide a GPON weak light fault positioning method and system based on a classification algorithm, and the method and system have the advantages of improving the fault positioning and scheduling processing efficiency.

The technical purpose of the invention is realized by the following technical scheme: a GPON weak light fault positioning method based on a classification algorithm comprises the following steps:

collecting and converging ONU optical power data;

constructing a training sample by combining optical power data of the ONU, network resource data and historical fault data;

constructing a GPON weak light fault positioning model based on a classification algorithm by using the training sample;

and analyzing the ONU optical power data to obtain fault section positioning data by applying the GPON weak light fault positioning model based on the classification algorithm.

Optionally, the acquiring and aggregating ONU optical power data includes:

starting an OLT periodic task to acquire the optical power of the ONU regularly, wherein the time interval of the OLT periodic task is recorded as T1;

collecting and synchronizing all the collected ONU optical power by taking a period T2 as a time interval; wherein T2> T1.

Optionally, the constructing a training sample by combining the ONU optical power data, the network resource data, and the historical fault data includes:

based on network resource data and ONU optical power data, calculating the ratio of weak light ONU in each T1 time period in all the first-stage optical splitters and the second-stage optical splitters, and recording as weak light characteristic data;

the network resource data is a topological connection relation among network elements in a GPON network, and comprises connection relations from an OLT to a first-stage optical splitter, connection relations from the first-stage optical splitter to a second-stage optical splitter, and connection relations from the second-stage optical splitter to ONUs, a main optical path between the OLT and the first-stage optical splitter is called a B section, a branch optical path between the first-stage optical splitter and the second-stage optical splitter is called a C1 section, and a branch optical path between the second-stage optical splitter and the ONUs is called a C2 section;

and combining the weak light characteristic data and the historical fault data to construct a training sample.

Optionally, the constructing a training sample by combining the weak light characteristic data and the historical fault data includes:

acquiring fault positioning information of historical fault data and corresponding fault time periods;

and constructing a training sample according to all fault positioning information and weak light characteristic data corresponding to the fault positioning information in the fault time period.

Optionally, the constructing a training sample according to all the fault location information and the weak light characteristic data in the corresponding fault time period includes:

when the fault positioning information is a fault of C2 section, calculating the ratio of weak light ONU under a first-level optical splitter and a second-level optical splitter respectively in the fault time section corresponding to the fault positioning information, and forming two training samples; one training sample is a first class-C2 abnormal-weak light ONU proportion under a secondary optical splitter; the other training sample is the weak light ONU proportion under a fourth class-C1 or C2 abnormal-first-level optical splitter;

when the fault positioning information is a fault of C1 section, calculating the ratio of weak light ONU under a first-level optical splitter and a second-level optical splitter respectively in the fault time section corresponding to the fault positioning information, and forming two training samples; one training sample is the ratio of weak light ONU under a second type-B or C1 abnormal-secondary optical splitter; the other training sample is the weak light ONU proportion under a fourth class-C1 or C2 abnormal-first-level optical splitter;

when the fault positioning information is a B-section fault, calculating the weak light ONU ratio under a first-stage optical splitter and a second-stage optical splitter respectively in a fault time section corresponding to the fault positioning information, and forming two training samples; one training sample is the ratio of weak light ONU under a second type-B or C1 abnormal-secondary optical splitter; the other training sample is the weak light ONU ratio under the third class-B abnormity-first-stage optical splitter.

Optionally, the constructing a GPON weak light fault location model based on a classification algorithm by using the training samples includes:

training and outputting a GPON weak light fault positioning model based on a classification algorithm by using a training sample;

the input of the GPON weak light fault positioning model based on the classification algorithm is the weak light ONU ratio under a first-level optical splitter and the weak light ONU ratio under a second-level optical splitter;

the GPON weak light fault location model based on the classification algorithm outputs fault section location data; the classification algorithm includes at least one of a support vector machine, a KNN algorithm, naive Bayes, a neural network, and a genetic algorithm.

Optionally, the classification algorithm uses a support vector machine algorithm, and an indirect method is adopted to construct a first classifier and a second classifier;

the input of the first classifier is the proportion of the second-level optical splitter ONU, the output of the first classifier is C2 section fault location information or C1/B section fault location information, and the first-class training samples and the second-class training samples are used for normalization processing and then input into a GPON weak light fault location model based on a classification algorithm for training; the input of the second classifier is the proportion of a first-level optical splitter ONU, the output of the second classifier is B-section fault positioning information or C1/C2-section fault positioning information, and after normalization processing is carried out by using training samples of a third category and a fourth category, the training samples are input into a GPON weak light fault positioning model based on a classification algorithm for training;

after calculation of training samples, two thresholds of a GPON weak light fault location model based on a classification algorithm are obtained, wherein the two thresholds comprise a first classifier secondary optical splitter weak light ONU duty ratio threshold TSH1 and a second classifier primary optical splitter weak light ONU duty ratio threshold TSH 2.

Optionally, the analyzing the ONU optical power data to obtain fault section positioning data by applying the classification algorithm-based GPON weak light fault positioning model includes:

calculating the ONU ratio of a first-level optical splitter and a second-level optical splitter of each T1 time period under each OLT according to the ONU optical power data;

when a weak light fault occurs, inquiring the ratio of weak light ONUs of a first-stage optical splitter and a second-stage optical splitter which the upper link of the fault belongs to and the ratio of weak light ONUs of the first-stage optical splitter and the second-stage optical splitter which the upper link of the fault belongs to in a time period of T1 which is the latest fault occurrence time;

and judging according to the weak light ONU ratio of the primary optical splitter and the secondary optical splitter to which the upper link of the fault belongs and the GPON weak light fault positioning model based on the classification algorithm, and outputting fault section positioning data.

Optionally, the determining according to the ratio of the weak light ONUs of the first-stage optical splitter and the second-stage optical splitter to which the fault upper link belongs by combining the GPON weak light fault location model includes:

judging whether the weak light ONU ratio of a secondary optical splitter to which the upper link of the fault belongs exceeds TSH1, and if not, judging that the C2 section of fault exists; if the number of the weak light ONU exceeds the number of the TSH2, judging that the weak light ONU ratio of the primary optical splitter to which the upper link of the fault belongs exceeds the number of the TSH2, and if the weak light ONU ratio does not exceed the number of the TSH2, judging that the C1 section of fault exists; if yes, judging that the section B is in fault.

A GPON weak light fault positioning system based on a classification algorithm comprises an acquisition module, a data preprocessing module, a classification algorithm model building module, an alarm preprocessing module and a human-computer interaction module;

the acquisition module is used for acquiring parameters such as an acquisition task period T1, a convergence synchronization period T2 and the like from the man-machine interaction module and acquiring optical power data of the ONU for the data preprocessing module to use; collecting resource topological data and historical fault data for a classification algorithm model building module to use; acquiring ONU weak light alarm data and resource topology data for an alarm preprocessing module to use;

the data preprocessing module is used for calculating the proportion of the weak light ONU of the first-level optical splitter and the proportion of the weak light ONU of the second-level optical splitter and carrying out normalized data preprocessing to obtain the proportion data of the weak light ONU and outputting the proportion data to the classification algorithm model building module and the alarm preprocessing module;

the classification algorithm model building module is used for acquiring resource topology data and historical fault data from the acquisition module, acquiring weak light ONU proportion data from the data preprocessing module and building training sample data; training a GPON weak light fault positioning model based on a classification algorithm through training sample data, and outputting the GPON weak light fault positioning model to an alarm preprocessing module for use; according to the instruction of the man-machine interaction module, periodically updating a GPON weak light fault positioning model based on a classification algorithm;

the alarm preprocessing module is used for acquiring ONU weak light alarm data and resource topology data from the acquisition module, acquiring a GPON weak light fault positioning model based on a classification algorithm from the classification algorithm model construction module, acquiring the first-stage splitter weak light ONU proportion and the second-stage splitter weak light ONU proportion at a specified time interval from the data preprocessing module, and preprocessing weak light alarms to output positioning reason classification;

and the human-computer interaction module is used for configuring parameters of an acquisition task period T1 and a convergence synchronization period T2 and controlling a GPON weak light fault positioning model based on a classification algorithm to retrain and update.

In conclusion, the invention has the following beneficial effects: acquiring ONU optical power data in a certain time period, and carrying out convergence synchronization on the ONU optical power data for subsequent operation and application; then combining ONU optical power data, network resource data in the GPON network and historical fault data to construct a training sample for training a classification algorithm, and outputting a GPON weak light fault positioning model based on the classification algorithm; when a GPON weak light fault occurs, analyzing ONU optical power data of a first-stage optical splitter and a second-stage optical splitter which are connected with a weak light ONU in an upper-link manner through a GPON weak light fault positioning model based on a classification algorithm, outputting fault section positioning data with the weak light fault, assigning a corresponding maintenance unit to the fault section according to the fault section positioning data, and repairing and processing the fault section by the corresponding maintenance unit.

Drawings

FIG. 1 is a schematic flow diagram of the process of the present invention;

figure 2 is a schematic diagram of a GPON network topology;

FIG. 3 is a graph of training sample data in the present invention;

FIG. 4 is a flowchart of GPON dim light fault location determination;

FIG. 5 is a block diagram of the modular components of the system of the present invention;

FIG. 6 is a schematic flow chart of the system of the present invention;

fig. 7 is an internal structural diagram of a computer device in an embodiment of the present invention.

Detailed Description

In order to make the objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below. Several embodiments of the invention are presented in the drawings. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein.

In the present invention, unless otherwise expressly specified or limited, the terms "mounted," "connected," "secured," and the like are to be construed broadly and can, for example, be fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations. The terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature.

In the present invention, unless expressly stated or limited otherwise, the recitation of a first feature "on" or "under" a second feature may include the recitation of the first and second features being in direct contact, and may also include the recitation that the first and second features are not in direct contact, but are in contact via another feature between them. Also, the first feature being "on," "above" and "over" the second feature includes the first feature being directly on and obliquely above the second feature, or merely indicating that the first feature is at a higher level than the second feature. A first feature being "under," "below," and "beneath" a second feature includes the first feature being directly under and obliquely below the second feature, or simply meaning that the first feature is at a lesser elevation than the second feature. The terms "vertical," "horizontal," "left," "right," "up," "down," and the like are used for descriptive purposes only and are not intended to indicate or imply that the referenced devices or elements must be in a particular orientation, configuration, and operation, and therefore should not be construed as limiting the present invention.

The invention is described in detail below with reference to the figures and examples.

The invention provides a GPON weak light fault positioning method and system based on a classification algorithm, as shown in figure 1, comprising:

step 100, collecting and converging ONU optical power data;

200, constructing a training sample by combining optical power data of the ONU, network resource data and historical fault data;

300, constructing a GPON weak light fault positioning model based on a classification algorithm by using the training sample;

and step 400, analyzing the ONU optical power data to obtain fault section positioning data by applying the GPON weak light fault positioning model based on the classification algorithm.

In practical application, optical power data of the ONU are collected in a certain time period, and are subjected to convergence synchronization for subsequent operation and application; then combining ONU optical power data, network resource data in the GPON network and historical fault data to construct a training sample for training a classification algorithm, and outputting a GPON weak light fault positioning model based on the classification algorithm; when a GPON weak light fault occurs, analyzing ONU optical power data of a first-stage optical splitter and a second-stage optical splitter which are connected with a weak light ONU in an upper-link manner through a GPON weak light fault positioning model based on a classification algorithm, outputting fault section positioning with the weak light fault, assigning an order to a maintenance unit corresponding to the fault section according to the fault section positioning, and repairing and processing the maintenance unit by the corresponding maintenance unit.

Further, the acquiring and aggregating ONU optical power data includes:

In practical application, the time interval T1 of the OLT periodic task and the synchronization period T2 for collecting all the ONU optical powers can be considered in combination with the requirements of real-time performance of fault handling and the cost of deployment operation, the length of T1 affects the granularity of the optical power index, the length of T2 affects the real-time performance of the optical power index, T1 can set 15-minute granularity or hour granularity, and T2 can set hour granularity or day granularity; and converging and synchronizing the collected optical power of all the ONUs into a common converging and synchronizing algorithm, such as a compression algorithm, a redundancy algorithm and the like.

the network resource data is a topological connection relation among network elements in a GPON network, and comprises connection relations from an OLT to a first-stage optical splitter, from the first-stage optical splitter to a second-stage optical splitter, and from the second-stage optical splitter to an ONU, a main optical path from the OLT to the first-stage optical splitter is called a section B, a branch optical path from the first-stage optical splitter to the second-stage optical splitter is called a section C1, and a branch optical path from the second-stage optical splitter to the ONU is called a section C2;

In practical application, on the basis of acquiring network resource topology connection relation data and acquired ONU optical power data, the weak-light ONU occupancy of each time segment (calculated at intervals of a period T1) in each primary optical splitter and each secondary optical splitter can be calculated, and the calculated weak-light ONU occupancy is used as a characteristic attribute of the positioning model. Preferably, the ONU weak light is defined as the ONU received optical power being lower than-27.0 dBm. The topological connection relationship between network elements in the GPON network is shown in fig. 2.

Optionally, the constructing a training sample by combining the low-light characteristic data and the historical fault data includes:

acquiring fault positioning information of historical fault data and a corresponding fault time period;

In practical application, the fault positioning information is obtained according to historical fault work order data to manually check and position fault section conclusions step by step; the selection of the fault time period can be based on the fault occurrence time in the historical fault work order, and the time period of the latest period T1 after the fault occurrence time is selected.

Further, the constructing a training sample according to all the fault location information and the weak light characteristic data in the corresponding fault time period includes:

when the fault positioning information is a fault of C2 section, calculating the ratio of weak light ONU under a first-level optical splitter and a second-level optical splitter respectively in the fault time section corresponding to the fault positioning information, and forming two training samples; one training sample is the ratio of weak light ONU under a first class-C2 abnormity-second-stage optical splitter; the other training sample is the weak light ONU proportion under a fourth class-C1 or C2 abnormal-first-level optical splitter;

when the fault positioning information is a B-section fault, calculating the weak light ONU ratio under a first-stage optical splitter and a second-stage optical splitter respectively in a fault time section corresponding to the fault positioning information, and forming two training samples; one training sample is the ratio of weak light ONU under a second type-B or C1 abnormal-secondary optical splitter; the other training sample is the weak light ONU proportion under the third class-B abnormity-first-stage optical splitter.

In practical application, according to the corresponding relationship and the fault location information, the duty ratio of the low-light ONU under the first-stage optical splitter and the duty ratio of the low-light ONU under the second-stage optical splitter under different faults are sampled, so as to obtain corresponding training sample data, which is shown in fig. 3.

Further, the constructing a GPON weak light fault location model based on a classification algorithm by using the training sample includes:

Further, the classification algorithm adopts a support vector machine algorithm and adopts an indirect method to construct a first classifier and a second classifier;

In practical application, a Support Vector Machine (SVM) algorithm is used as a model algorithm, the SVM mainly solves the problem of binary classification, and the ONU weak light fault section positioning data includes B, C1 and C2, and 2 classifiers can be constructed by an indirect method to solve the problem of multi-class classification. Selecting 60 groups of training samples of various categories for training respectively; the model thresholds TSH1 and TSH2 may vary with the network size, and should be retrained and updated at a certain period, such as calculation and update every month or every quarter, to improve the application effect of the model.

Further, as shown in fig. 4, the analyzing the ONU optical power data to obtain fault section positioning data by applying the classification algorithm-based GPON weak light fault positioning model includes:

when a weak light fault occurs, inquiring the first-stage optical splitter and the second-stage optical splitter which the fault upper link belongs to and the weak light ONU ratio of the first-stage optical splitter and the second-stage optical splitter which the fault upper link belongs to in a time period of T1 closest to the fault occurrence time;

and judging according to the weak light ONU ratio of the primary optical splitter and the secondary optical splitter which the fault upper link belongs to by combining the GPON weak light fault positioning model based on the classification algorithm, and outputting fault section positioning data.

Further, the determining according to the ratio of the weak light ONUs of the first-stage optical splitter and the second-stage optical splitter to which the upper link of the fault belongs and by combining a GPON weak light fault positioning model includes:

In practical application, the C2 section deals with the problem that the optical power of the ONU optical module does not reach the standard or the distance between the ONU and the secondary optical splitter exceeds the standard value; the C1 section deals with the loss problem of the branch optical fiber circuit or the secondary optical splitter; trunk fiber line loss, first-level optical splitter loss or PON mouth optical module trouble scheduling problem are handled to B section, and the weak light ONU of the first-level optical splitter that allies oneself with on the trouble belonged to and second-level optical splitter accounts for the ratio and carries out the analysis and judge, can obtain accurate trouble section positioning data to can directly distribute a bill to corresponding trouble section maintenance team according to this trouble section positioning data, avoid distributing a bill simultaneously and carry out the investigation step by step for a plurality of teams, improve location treatment effeciency.

As shown in fig. 5, the present invention further provides a GPON weak light fault location system based on a classification algorithm, including: the system comprises an acquisition module 10, a data preprocessing module 20, a classification algorithm model construction module 30, an alarm preprocessing module 40, an intelligent order dispatching module 50 and a human-computer interaction module 60;

the acquisition module 10 is configured to acquire parameters such as an acquisition task period T1 and a convergence synchronization period T2 from the human-computer interaction module, and acquire optical power data of the ONU for the data preprocessing module to use; collecting resource topological data and historical fault data for a classification algorithm model building module to use; acquiring ONU weak light alarm data and resource topology data for an alarm preprocessing module to use;

the data preprocessing module 20 is configured to calculate a first-stage splitter weak-light ONU proportion and a second-stage splitter weak-light ONU proportion and perform normalized data preprocessing to obtain weak-light ONU proportion data, and output the weak-light ONU proportion data to the classification algorithm model building module and the alarm preprocessing module;

the classification algorithm model building module 30 is configured to obtain resource topology data and historical fault data from the acquisition module, obtain weak light ONU proportion data from the data preprocessing module, and build training sample data; training a GPON weak light fault positioning model based on a classification algorithm through training sample data, and outputting the GPON weak light fault positioning model to an alarm preprocessing module for use; according to the instruction of the man-machine interaction module, periodically updating a GPON weak light fault positioning model based on a classification algorithm;

the alarm preprocessing module 40 is used for acquiring ONU weak light alarm data and resource topology data from the acquisition module, acquiring a GPON weak light fault location model based on a classification algorithm from the classification algorithm model construction module, acquiring a primary splitter weak light ONU proportion and a secondary splitter weak light ONU proportion at a specified time period from the data preprocessing module, preprocessing weak light alarms, outputting location reason classification, and outputting a location result to the intelligent dispatch module;

the intelligent dispatching module 50 is used for receiving the preprocessed alarm data and the preliminary positioning result, and dispatching the fault work order to the corresponding maintenance team for processing;

the human-computer interaction module 60 is configured to configure acquisition task period T1 and convergence synchronization period T2 parameters, and control the GPON weak light fault location model based on the classification algorithm to retrain and update.

Specifically, as shown in fig. 6, the acquisition module acquires data such as optical power data of each ONU and ONU weak light alarm from the OLT, acquires topology resource relationship data from the resource management system, and acquires historical fault data from the work order system. And the other modules acquire data from the acquisition module in a pushing mode of the acquisition module or in an acquisition mode of other modules according to requirements.

The data preprocessing module acquires ONU optical power data and resource data, and calculates the weak light ONU ratio of the primary and secondary optical splitters of each T1 under each OLT. And judging whether the fault positioning model needs to be updated or whether weak light fault alarm needs to be processed or not, and if yes, pushing the preprocessed data in the corresponding time period to a classification algorithm model building module or an alarm preprocessing module.

The classification model construction module acquires model construction and updating instructions from the man-machine interaction module, forms training samples according to resource data, historical fault data and weak light ONU proportion data, and further constructs a GPON weak light fault positioning model based on a classification algorithm for the alarm preprocessing module to use.

And the alarm preprocessing module detects whether a weak light fault alarm to be processed exists, if so, weak light ONU proportion data of the optical splitter and the corresponding time period are obtained, and a latest GPON weak light fault positioning model is obtained, so that intelligent fault positioning and reason classification are realized. And pushing the fault alarm and positioning result to an intelligent dispatching module to dispatch to a corresponding maintenance team. The latest GPON weak light fault positioning model is obtained, and preferably, the latest GPON weak light fault positioning model can be pushed to an alarm preprocessing module in an active pushing mode after a classification model building module is updated.

The man-machine interaction module can set parameters such as an ONU optical power acquisition task period T1 and a convergence synchronization period T2 and perform instruction operation of updating a GPON weak light fault positioning model.

For specific limitations of a GPON low-light fault location system based on a classification algorithm, reference may be made to the above limitations of a GPON low-light fault location method and system based on a classification algorithm, which are not described herein again. All modules in the GPON weak light fault positioning system based on the classification algorithm can be completely or partially realized through software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.

In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 7. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The computer program is executed by a processor to implement a GPON dim light fault location method based on a classification algorithm.

Those skilled in the art will appreciate that the architecture shown in fig. 7 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:

collecting and converging ONU optical power data;

combining the ONU optical power data, the network resource data and the historical fault data to construct a training sample;

In one embodiment, the collecting and aggregating ONU optical power data comprises:

In one embodiment, the constructing a training sample in combination with the ONU optical power data, the network resource data, and the historical failure data includes:

In one embodiment, the constructing a training sample by combining the low-light characteristic data and the historical fault data comprises:

In one embodiment, the constructing a training sample according to all the fault location information and the weak light characteristic data in the corresponding fault time period includes:

In one embodiment, the constructing a GPON weak light fault location model based on a classification algorithm by using the training samples includes:

In one embodiment, the classification algorithm adopts a support vector machine algorithm, and a first classifier and a second classifier are constructed by adopting an indirect method;

In an embodiment, the analyzing the ONU optical power data to obtain fault section positioning data by applying the classification algorithm-based GPON weak light fault positioning model includes:

In one embodiment, the determining according to the ratio of weak light ONUs of the first optical splitter and the second optical splitter to which the fault upper link belongs in combination with a GPON weak light fault location model includes:

judging whether the duty ratio of a weak light ONU of a secondary optical splitter to which the upper link of the fault belongs exceeds TSH1, and if not, judging that the fault is at a C2 section; if the number of the weak light ONU exceeds the number of the TSH2, judging that the weak light ONU ratio of the primary optical splitter to which the upper link of the fault belongs exceeds the number of the TSH2, and if the weak light ONU ratio does not exceed the number of the TSH2, judging that the C1 section of fault exists; if yes, judging that the section B is in fault.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database or other medium used in the embodiments provided herein can include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above description is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments, and all technical solutions belonging to the idea of the present invention belong to the protection scope of the present invention. It should be noted that modifications and embellishments within the scope of the invention may occur to those skilled in the art without departing from the principle of the invention, and are considered to be within the scope of the invention.

Claims

1. A GPON weak light fault positioning method based on a classification algorithm is characterized by comprising the following steps:

collecting and converging ONU optical power data;

2. The method of claim 1, wherein collecting and aggregating ONU optical power data comprises:

3. The method of claim 2, wherein the constructing training samples in combination with the ONU optical power data, the network resource data, and the historical failure data comprises:

4. The method of claim 3, wherein the combining the low-light signature data and the historical fault data to construct training samples comprises:

5. The method according to claim 4, wherein the constructing the training sample according to all the fault location information and the dim light characteristic data at the corresponding fault time period comprises:

6. The method of claim 5, wherein the constructing a GPON weak light fault location model based on a classification algorithm by using the training samples comprises:

7. The method according to claim 6, wherein the classification algorithm is a support vector machine algorithm, and the first classifier and the second classifier are constructed by an indirect method;

8. The method of claim 7, wherein the applying the classification algorithm-based GPON weak light fault location model to analyze the ONU optical power data to obtain fault section location data comprises:

9. The method according to claim 8, wherein the determining according to the ratio of weak light ONUs of the first splitter and the second splitter to which the fault upper link belongs in combination with a GPON weak light fault location model includes:

10. A GPON weak light fault positioning system based on a classification algorithm is characterized by comprising the following components:

the system comprises an acquisition module, a data preprocessing module, a classification algorithm model building module, an alarm preprocessing module and a human-computer interaction module;

the data preprocessing module is used for calculating the proportion of the weak light ONU of the first-stage optical splitter and the proportion of the weak light ONU of the second-stage optical splitter and carrying out normalization data preprocessing to obtain proportion data of the weak light ONU and output the proportion data to the classification algorithm model building module and the alarm preprocessing module;